High capacity cascade-type mineral sorting machine and method

Abstract

Methods and systems for achieving higher efficiencies and capacities in sorting feed material are described herein, such as for separating desirable “good” rock or ore from undesirable “bad” rock or ore in an unsegregated, unseparated stream of feed material. In the disclosure, higher efficiencies are achieved with combinations of multiple sensor/diverter cells in stages in a cascade arrangement. The number and combination of cells in the cascade may be determined through a priori characterization of probabilities involved in sensor/rock and rock/diverter interactions, and mathematical determinations of the optimal number and combination of stages based on this probability. Further, as disclosed herein, desired sorting capacities are achieved through addition of multiple cascades in parallel until the desired sorting capacity is reached.

We claim: 
     
       1. A system for sorting ore from a stream of material, comprising:
 a first size-classifying stage configured to separate at least a portion of the stream of material into at least fine fractions and coarse fractions; 
 a first sorting cascade comprising at least one sorting cell, wherein the first sorting cascade is configured to:
 receive the coarse fractions; 
 detect spectral characteristics of mineral samples of the coarse fractions as the mineral samples move past electromagnetic sensors; 
 identify a composition of each of the mineral samples by comparing the spectral characteristics of the individual mineral samples with spectral characteristics of known mineral samples using a trained conditional random field-type pattern matching algorithm with nearest neighbor detection, wherein the composition includes identification of multiple mineral constituents; 
 based on the composition of each mineral sample, detect content of at least a first desired component from the coarse fractions; and 
 sort, based on a first grade threshold, the coarse fractions into a coarse fraction accept stream and a coarse fraction reject stream; and 
 
 a second sorting cascade comprising at least one sorting cell, wherein the second sorting cascade is configured to:
 receive the fine fractions; 
 detect content of at least a second desired component from the fine fractions; and 
 sort, based on a second grade threshold, the fine fractions into a fine fraction accept stream and a fine fraction reject stream. 
 
 
     
     
       2. The system of  claim 1 , wherein detecting the content of at least the first desired component and detecting the content of at least the second desired component includes detecting the content of a same desired component in the fine fractions and the coarse fractions. 
     
     
       3. The system of  claim 2 , wherein the first grade threshold is different from the second grade threshold. 
     
     
       4. The system of  claim 1 , wherein detecting the content of at least the first desired component and detecting the content of at least the second desired component includes detecting a content of a different desired component in the fine fractions and the coarse fractions. 
     
     
       5. The system of  claim 1 , further comprising a second size-classifying stage configured to separate at least a portion of the coarse fractions into at least second-stage fine fractions and second-stage coarse fractions. 
     
     
       6. The system of  claim 1 , wherein a number of sorting cells in the first sorting cascade is determined by:
 calculating a probability of correctly determining the content of the first desired component of the coarse fractions using a sensor; 
 calculating a probability of correctly diverting the coarse fractions using a diverter; 
 calculating a utility of the first sorting cascade based on the probability of correctly determining the content of the first desired component of the coarse fractions and the probability of correctly diverting the coarse fractions; and 
 determining the number of the at least one sorting cell in the first sorting cascade based on the calculated utility. 
 
     
     
       7. The system of  claim 6 , further comprising at least one additional sorting cascade, wherein each additional sorting cascade comprises the determined number of the at least one sorting cell. 
     
     
       8. The system of  claim 7 , wherein a number of additional sorting cascades is determined based on a desired separation capacity. 
     
     
       9. A system for sorting ore from a stream of material, comprising:
 a first size-classifying stage configured to separate at least a portion of the stream of material into at least fine fractions and coarse fractions; 
 a first sorting cascade comprising at least one sorting cell, wherein the first sorting cascade is configured to:
 receive the coarse fractions; 
 detect spectral characteristics of mineral samples of the course coarse fractions as the mineral samples move past electromagnetic sensors; 
 identify a composition of each of the mineral samples by comparing the spectral characteristics of the individual mineral samples with spectral characteristics of known mineral samples using a trained conditional random field-type pattern matching algorithm with nearest neighbor detection, wherein the composition includes identification of multiple mineral constituents; 
 based on the composition of each mineral sample, detect content of at least a first desired component from the coarse fractions; and 
 sort, based on a first grade threshold, the coarse fractions into a coarse fraction accept stream and a coarse fraction reject stream; 
 
 a second sorting cascade comprising at least one sorting cell, wherein the second sorting cascade is configured to:
 receive the fine fractions; 
 detect content of at least a second desired component from the fine fractions; and 
 sort, based on a second grade threshold, the fine fractions into a fine fraction accept stream and a fine fraction reject stream; and 
 
 an apparatus to provide a product stream comprising the fine fraction accept stream and the coarse fraction accept stream. 
 
     
     
       10. The system of  claim 9 , wherein detecting the content of at least the first desired component and detecting the content of at least the second desired component includes detecting the content of a same desired component in the fine fractions and the coarse fractions, and wherein the first grade threshold is different from the second grade threshold. 
     
     
       11. The system of  claim 9 , wherein detecting the content of at least the first desired component and detecting the content of at least the second desired component includes detecting a content of a different desired component in the fine fractions and the coarse fractions. 
     
     
       12. The system of  claim 9 , further comprising a second size-classifying stage configured to separate at least a portion of the coarse fractions into at least second-stage fine fractions and second-stage coarse fractions. 
     
     
       13. The system of  claim 9 , wherein a number of sorting cells in the first sorting cascade is determined by:
 determining a probability of correctly determining the content of the first desired component of the coarse fractions using a sensor; 
 determining a probability of correctly diverting the coarse fractions using a diverter; 
 determining a utility of the first sorting cascade based on the probability of correctly determining the content of the first desired component of the coarse fractions and the probability of correctly diverting the coarse fractions; and 
 determining the number of the at least one sorting cell in the first sorting cascade based on the determined utility; 
 and wherein the system further comprises at least one additional sorting cascade, wherein each additional sorting cascade comprises the determined number of the at least one sorting cell, and wherein a number of additional sorting cascades is determined based on a desired separation capacity. 
 
     
     
       14. The system of  claim 9 , wherein a number of sorting cells in the first sorting cascade is determined by:
 determining a probability of correctly determining the content of the first desired component of the coarse fractions using a sensor; 
 determining a probability of correctly diverting the coarse fractions using a diverter; 
 determining a utility of the first sorting cascade based on the probability of correctly determining the content of the first desired component of the coarse fractions and the probability of correctly diverting the coarse fractions; and 
 determining the number of the at least one sorting cell in the first sorting cascade based on the determined utility. 
 
     
     
       15. A system for sorting ore from a stream of material, comprising:
 first size-classifying means configured to separate at least a portion of the stream of material into at least fine fractions and coarse fractions; 
 first sorting means configured to:
 receive the coarse fractions; 
 detect spectral characteristics of mineral samples of the coarse fractions as the mineral samples move past electromagnetic sensors; 
 identify a composition each of the mineral samples by comparing the spectral characteristics of the individual mineral samples with spectral characteristics of known mineral samples using a trained conditional random field-type pattern matching algorithm with nearest neighbor detection, wherein the composition includes identification of multiple mineral constituents; 
 based on the composition of each mineral sample, detect content of at least a first desired component from the coarse fractions; and 
 sort, based on a first grade threshold, the coarse fractions into a coarse fraction accept stream and a coarse fraction reject stream; 
 
 a second sorting means configured to:
 receive the fine fractions; 
 detect content of at least a second desired component from the fine fractions; and 
 sort, based on a second grade threshold, the fine fractions into a fine fraction accept stream and a fine fraction reject stream; and 
 
 means for providing a product stream comprising the fine fraction accept stream and the coarse fraction accept stream. 
 
     
     
       16. The system of  claim 15 , wherein detecting the content of at least the first desired component and detecting the content of at least the second desired component includes detecting the content of a same desired component in the fine fractions and the coarse fractions, and wherein the first grade threshold is different from the second grade threshold. 
     
     
       17. The system of  claim 15 , wherein detecting the content of at least the first desired component and detecting the content of at least the second desired component includes detecting a content of a different desired component in the fine fractions and the coarse fractions. 
     
     
       18. The system of  claim 15 , further comprising second size-classifying means configured to separate at least a portion of the coarse fractions into at least second-stage fine fractions and second-stage coarse fractions. 
     
     
       19. The system of  claim 15 , wherein the first sorting means includes a number of sorting cells determined based on:
 a probability of correctly determining the content of the first desired component of the coarse fractions using a sensor; 
 a probability of correctly diverting the coarse fractions using a diverter; 
 a utility of the first sorting cascade based on the probability of correctly determining the content of the first desired component of the coarse fractions and the probability of correctly diverting the coarse fractions; and 
 wherein the system further comprises at least one additional sorting means, wherein each additional sorting means comprises the determined number of the at least one sorting cell, and wherein a number of additional sorting means is determined based on a desired separation capacity.